Copy-Paste Detection Based on a SIFT Marked Graph Feature Vector

To detect copy-paste tampering, an im-proved SIFT (Scale invariant feature transform)-based al-gorithm was proposed. Maximum angle is defined and a maximum angle-based marked graph is constructed. The m arked graph feature vector is provided to each SIFT key point via discrete polar coordinate transformation. Key points are matched to detect the copy-paste tamper-ing regions. The experimental results show that the pro-posed algorithm can effectively identify and detect the ro-tated or scaled copy-paste regions, and in comparison with the methods reported previously, it is resistant to post-processing, such as blurring, Gaussian white noise and JPEG recompression. The proposed algorithm performs better than the existing algorithm to dealing with scaling transformation.     

Weighted archetypal analysis of the multi-element graph for query-focused multi-document summarization

Most existing research on applying the matrix factorization approaches to query-focused multi-document summarization (Q-MDS) explores either soft/hard clustering or low rank approximation methods. We employ a different kind of matrix factorization method, namely weighted archetypal analysis (wAA) to Q-MDS. In query-focused summarization, given a graph representation of a set of sentences weighted by similarity to the given query, positively and/or negatively salient sentences are values on the weighted data set boundary. We choose to use wAA to compute these extreme values, archetypes, and hence to estimate the importance of sentences in target documents set. We investigate the impact of using the multi-element graph model for query focused summarization via wAA. We conducted experiments on the data of document understanding conference (DUC) 2005 and 2006. Experimental results evidence the improvement of the proposed approach over other closely related methods and many of state-of-the-art systems.     

The impact of study load on the dynamics of longitudinal email communications among students

With the advent of information technology, emails have gained wide acceptability among students as an asynchronous communication tool. According to the current pedagogy literature the overall trend of the use of email communication by university students has been increasing significantly since its inception, despite the rapid growth of the popularity and acceptability of other social mediums (e.g. Mobile phone and Facebook). In this study, we explore a longitudinal email communication network, which evolved under an increasing study load among 38 students throughout a university semester, using measures of social network analysis (SNA) and exponential random graph (ERG) models. This longitudinal network was divided into three waves, where each wave represents the portion of the complete longitudinal network that evolves between two consecutive observations. An increased study load was imposed through the assessment components of the course. SNA measures of degree centrality (i.e. the activity of an actor or actor popularity), betweenness centrality (i.e. the capacity to control the flow of information in a network), closeness centrality (i.e. reachable to other nodes) and reciprocity (i.e. tendency to make reciprocal links) are considered to explore this longitudinal network. ERG models are probabilistic models that are presented by locally determined explanatory variables and can effectively identify structural properties of networks. From the analysis of this email communication network, we notice that students' network positions and behaviours change with the changes in their study load. In particular, we find that (i) students make an increased number of email communications with the increase of study load; (ii) the email communication network become sparse with the increase of study load; and (iii) the 2-star parameter (a subset of three nodes in which one node is connected to each of the other two nodes) and the triangle parameter (a subset of three nodes in which each node is connected to the other two nodes) can effectively explain the formation of network in wave3; whereas, the 3-star parameter (a subset of four nodes in which one node is connected to each of other three nodes) can effectively explain the formation of network in wave1 and wave2. Interpretations of these findings for the monitoring of student behaviour in online learning environments, as well as the implications for the design of assessment and the use of asynchronous tools are discussed in this paper.     

Efficient algorithm for calculating short cycles in Tanner graph based on matrix computation

Loop distribution of Tanner graph affects the BER performance of low-density parity-check codes(LDPC) decoding.To count short cycles in the Tanner graph efficiently,a side by side recursion algorithm based on matrix computation was proposed.Firstly,5 basic graph structures were defined to realize recursive calculate in the implementation process.Compared with previous works,the algorithm provided many methods for counting the same length of cycles.The same result confirmed the correctness of the algorithm.The new algorithm could not only calculate the total number of cycles,but also gave the number each edge participating in fixed-length cycles.Its complexity was proportional to the product of D and square of N,where D was the average degree of variable nodes,and N denoted the code length.For LDPC codes,D was far less than N.For most of the LDPC codes,the calculation for numbers of cycle-length g、g+2、g+4 was only several seconds.     

A distributed command governor based on graph colorability theory

In this paper, a distributed command governor (CG) strategy is introduced that, by the use of graph colorability theory, improves the scalability property and the performance of recently introduced distributed noncooperative sequential CG strategies. The latter are characterized by the fact that only 1 agent at a decision time is allowed to update its command, whereas all the others keep applying their previously computed commands. The scalability of these early CG distributed schemes and their performance are limited because the structure of the constraints is not taken into account in their implementation. Here, by exploiting the idea that agents that are not directly coupled by the constraints can simultaneously update their control actions, the agents in the network are grouped into particular subsets (turns). At each time instant, on the basis of a round‐robin policy, all agents belonging to a turn are allowed to update simultaneously their commands, whereas agents in other turns keep applying their previous commands. Then, a turn‐based distributed CG strategy is proposed and its main properties are analyzed. Graph colorability theory is used to determine the minimal number of turns and to distribute each agent in at least a turn. A novel graph colorability problem that allows one to maximize the frequency at which agents can update their commands is proposed and discussed. A final example is presented to illustrate the effectiveness of the proposed strategy.     

基于二分图的宝鸡公交线路-站点网络建模及分析

复杂网络理论已成为研究交通系统的有效工具。以陕西省宝鸡市的43个公交线路及344个公交站点数据为基础,基于二分图理论,构建了站点-线路公交网络模型,并对该网络的统计性质进行了分析。研究结果表明,网络中的线路节点度分布呈现泊松分布,大部分线路度值集中在平均值附近;而站点节点度分布呈现幂律分布,其中的少数节点高度连接,为网络中的关键枢纽节点。这些高度连接的线路和站点可被筛选为宝鸡市交通网络研究中的重点规划对象。     

Supporting ontology adaptation and versioning based on a graph of relevance

Ontologies recently have become a topic of interest in computer science since they are seen as a semantic support to explicit and enrich data-models as well as to ensure interoperability of data. Moreover, supporting ontology adaptation becomes essential and extremely important, mainly when using ontologies in changing environments. An important issue when dealing with ontology adaptation is the management of several versions. Ontology versioning is a complex and multifaceted problem as it should take into account change management, versions storage and access, consistency issues, etc. The purpose of this paper is to propose an approach and tool for ontology adaptation and versioning. A series of techniques are proposed to ‘safely’ evolve a given ontology and produce a new consistent version. The ontology versions are ordered in a graph according to their relevance. The relevance is computed based on four criteria: conceptualisation, usage frequency, abstraction and completeness. The techniques to carry out the versioning process are implemented in the Consistology tool, which has been developed to assist users in expressing adaptation requirements and managing ontology versions.     

Hierarchical Implicit Shape Modeling

In this paper, a new hierarchical approach for object detection is proposed. Object detection methods based on Implicit Shape Model (ISM) efficiently handle deformable objects, occlusions and clutters. The structure of each object in ISM is defined by a spring like graph. We introduce hierarchical ISM in which structure of each object is defined by a hierarchical star graph. Hierarchical ISM has two layers. In the first layer, a set of local ISMs are used to model object parts. In the second layer, structure of parts with respect to the object center is modeled by global ISM. In the proposed approach, the obtained parts for each object category have high discriminative ability. Therefore, our approach does not require a verification stage. We applied the proposed approach to some datasets and compared the performance of our algorithm to comparable methods. The results show that our method has a superior performance.     

Graph-based Lexicalized Reordering Models for Statistical Machine Translation

Lexicalized reordering models are very important components of phrasebased translation systems.By examining the reordering relationships between adjacent phrases,conventional methods learn these models from the word aligned bilingual corpus,while ignoring the effect of the number of adjacent bilingual phrases.In this paper,we propose a method to take the number of adjacent phrases into account for better estimation of reordering models.Instead of just checking whether there is one phrase adjacent to a given phrase,our method firstly uses a compact structure named reordering graph to represent all phrase segmentations of a parallel sentence,then the effect of the adjacent phrase number can be quantified in a forward-backward fashion,and finally incorporated into the estimation of reordering models.Experimental results on the NIST Chinese-English and WMT French-Spanish data sets show that our approach significantly outperforms the baseline method.